U.S. patent application number 10/667477 was filed with the patent office on 2004-06-17 for method for imaging in the computer tomography of a periodically moved object to be examined and ct device for carrying out the method.
Invention is credited to Bruder, Herbert, Flohr, Thomas, Stierstorfer, Karl.
Application Number | 20040114708 10/667477 |
Document ID | / |
Family ID | 31983999 |
Filed Date | 2004-06-17 |
United States Patent
Application |
20040114708 |
Kind Code |
A1 |
Bruder, Herbert ; et
al. |
June 17, 2004 |
Method for imaging in the computer tomography of a periodically
moved object to be examined and CT device for carrying out the
method
Abstract
A method is for imaging in computer tomography, in which a
periodically moved object to be examined is scanned with the aid of
a beam of rays coming from a focus. A two-dimensionally designed
detector array generates output data which are filtered in a
suitable way and back-projected in order to obtain at least one
sectional image which represents the absorption values of the
section of the object to be examined in a particular movement
state. In the method, on the one hand, a weighting function which
weights the spatial distance of a ray in question from the voxel in
question is used for the back-projection. Further, on the other
hand, a weighting function which represents the time difference
from the examination area movement state to be represented in each
case is also used.
Inventors: |
Bruder, Herbert;
(Hoechstadt, DE) ; Flohr, Thomas; (Uehlfeld,
DE) ; Stierstorfer, Karl; (Erlangen, DE) |
Correspondence
Address: |
HARNESS, DICKEY & PIERCE, P.L.C.
P.O.BOX 8910
RESTON
VA
20195
US
|
Family ID: |
31983999 |
Appl. No.: |
10/667477 |
Filed: |
September 23, 2003 |
Current U.S.
Class: |
378/4 |
Current CPC
Class: |
A61B 6/027 20130101;
A61B 6/4028 20130101; G01N 2223/419 20130101; Y10S 378/901
20130101; A61B 6/503 20130101; A61B 6/541 20130101; G01N 23/046
20130101; A61B 6/4085 20130101; A61B 6/032 20130101 |
Class at
Publication: |
378/004 |
International
Class: |
G01N 023/00; A61B
006/00; H05G 001/60; G21K 001/12 |
Foreign Application Data
Date |
Code |
Application Number |
Sep 23, 2002 |
DE |
10244180.4 |
Claims
What is claimed is:
1. A method for generating CT images of a periodically moving
examination area, comprising: moving at least one focus, in order
to scan a periodically movable object to be examined with a beam of
rays coming from the at least one focus and with a
two-dimensionally designed detector array with a multiplicity of
distributed detector elements for detecting the rays of the beam of
rays, relative to the object to be examined on at least one focal
track circumscribing the object to be examined with the detector
array lying opposite, wherein the detector elements of the detector
array are adapted to deliver output data representing an
attenuation of the rays when passed through the periodically
movable object; simultaneously collecting movement data upon
movement of the movable object, to enable assigning of detector
data and data resulting therefrom to movement states; filtering the
output data; three-dimensionally back-projecting the filtered
output data to generate at least one sectional image of a layer of
the object to be examined having a layer thickness, wherein each
sectional image represents absorption values, obtained from the
output data, of the voxel belonging to the layer of the object to
be examined for the radiation of the beam of rays, wherein a
weighting function, for weighting the spatial distance of a ray in
question from the voxel in question, is used for the
back-projecting, and wherein a weighting function, representing a
time difference from the examination area movement state to be
represented is also used for the back-projecting.
2. The method as claimed in claim 1, wherein the filtering is
carried out in the direction of a tangent to the focal track
belonging to the respective focal position.
3. The method as claimed in claim 1, wherein a conversion of the
output data, obtained in the form of rays P(.alpha.,.beta.,q) in a
fan-ray geometry into parallel data available in the form of at
least one of rays P(.theta., .beta.,q) and P(.theta.,p,q) in a
parallel-ray geometry, is carried out before the filtering, wherein
.alpha. is the focal angle, .beta. is the fan angle, q is the row
index of the detector system corresponding to the z coordinate,
.theta.=.alpha.+.beta. is the parallel fan angle, p=R.sub.F
sin(.beta.) is the parallel coordinate corresponding to the ray
distance from the rotation axis (system axis), and R.sub.F is the
radius of the focal track.
4. The method as claimed in claim 3, wherein the back-projecting of
the parallel data is carried out in such a way that the sum 12 P x
, y , z ( ) = k q h z ( d x , y , z ( + k , { p ~ ~ } , q ) ) h
phase ( t ( + k ) - c R ( k ) ) P ( + k , { p ~ ~ } , q ) is formed
in the course of the back-projecting for each voxel (x,y,z) for
each .theta..di-elect cons.[0,.pi.[ for the rays
P(.theta.+k.pi.,{tilde over (.beta.)}, q) or P(.theta.+k.pi.,{tilde
over (p)}, q) whose projection along the system axis passes through
(x,y), wherein x,y,z are the coordinates of the respective voxel, k
is an integer corresponding to the number of half-circuits of the
focus which are involved in the reconstruction, {tilde over (p)}
are the parallel coordinates of those rays whose projections along
the system axis pass through the coordinates (x,y) of the
respective voxel (x,y,z), {tilde over (.beta.)} are the fan angles
of those rays whose projections along the system axis pass through
the coordinates (x,y) of the respective voxel (x,y,z), h.sub.z is a
weighting function determining the layer thickness of the layer of
the object to be examined which is represented in the sectional
image being produced, d is a function which is equal to the
distance of the respective ray from the corresponding voxel (x,y,z)
or depends on the distance of the respective ray from the
corresponding voxel (x,y,z), and h.sub.phase is a weighting
function relating to the time difference of the projection
P(.theta.+k.pi.,{tilde over (.beta.)}, q) or P(.theta.+k.pi.,
{tilde over (p)}q) from a movement state, C.sub.R(k) representing a
time position which determines the periodic movement state of an
examination area, preferably an assigned heart cycle.
5. The method as claimed in claim 4, wherein the sum 13 P x , y , z
( ) = 1 H k q h z ( d x , y , z ( + k , { p ~ ~ } , q ) ) h phase (
t ( + k ) - c R ( k ) ) P ( + k , { p ~ ~ } , q ) which is
normalized to the sum H of the weights h 14 H = k q h z ( d x , y ,
z ( + k , { p ~ ~ } , q ) ) h phase ( t ( + k ) - c R ( k ) ) is
formed during the back-projection of the parallel data.
6. The method as claimed in claim 1, wherein the rays of each voxel
in question are weighted during the back-projecting in such a way
that rays striking the detector elements centrally are weighted
relatively more heavily than rays striking at the edge.
7. The method as claimed in claim 1, wherein the focal track is a
circular track.
8. The method as claimed in claim 1, wherein the focal track is a
spiral track, obtained by moving the focus on the circular track
about the system axis and simultaneously carrying out a relative
movement between the focus and the object to be examined in the
direction of the system axis.
9. The method as claimed in claim 1, wherein the detector elements
are arranged distributed in rows and columns on the detector
array.
10. A CT device for scanning a periodically movable object to be
examined, with a beam of rays coming from at least one focus and
with a two-dimensionally designed detector array with a
multiplicity of distributed detector elements for detecting the
rays of the beam of rays, the at least one focus being moved
relative to the object to be examined on at least one focal track
circumscribing the object to be examined with the detector array
lying opposite, the CT device including means for detection of the
movement state of the object to be examined, for collection of the
detector data, for filtering and back-projection as claimed in
claim 1.
11. The CT device as claimed in claim 10, wherein each of said
means are implemented at least partially by at least one of
programs and program modules.
12. The method of claim 1, wherein the method is for generating CT
images of a heart region of a living being.
13. The method of claim 1, wherein the method is for generating CT
images of a heart region of a patient.
14. The method as claimed in claim 2, wherein a conversion of the
output data, obtained in the form of rays P(.alpha., .beta., q) in
a fan-ray geometry into parallel data available in the form of at
least one of rays P(.theta., .beta., q) and P(.theta., p, q) in a
parallel-ray geometry, is carried out before the filtering, wherein
.alpha. is the focal angle, .beta. is the fan angle, q is the row
index of the detector system corresponding to the z coordinate,
.theta.=.beta.+.beta. is the parallel fan angle, p=R.sub.F
sin(.beta.) is the parallel coordinate corresponding to the ray
distance from the rotation axis (system axis), and R.sub.F is the
radius of the focal track.
15. The method as claimed in claim 14, wherein the back-projecting
of the parallel data is carried out in such a way that the sum 15 P
x , y , z ( ) = k q h z ( d x , y , z ( + k , { p ~ ~ } , q ) ) h
phase ( t ( + k ) - c R ( k ) ) P ( + k , { p ~ ~ } , q ) is formed
in the course of the back-projecting for each voxel (x, y, z) for
each a .theta..di-elect cons.[0,.pi.r[ for the rays
P(.theta.+k.pi., {tilde over (.beta.)}, q) or P(.theta.+k.pi.,
{tilde over (p)}, q) whose projection along the system axis passes
through (x,y), wherein x, y, z are the coordinates of the
respective voxel, k is an integer corresponding to the number of
half-circuits of the focus which are involved in the
reconstruction, {tilde over (p)} are the parallel coordinates of
those rays whose projections along the system axis pass through the
coordinates (x, y) of the respective voxel (x, y, z), {tilde over
(.beta.)} are the fan angles of those rays whose projections along
the system axis pass through the coordinates (x, y) of the
respective voxel (x, y, z), h.sub.z is a weighting function
determining the layer thickness of the layer of the object to be
examined which is represented in the sectional image being
produced, d is a function which is equal to the distance of the
respective ray from the corresponding voxel (x, y, z) or depends on
the distance of the respective ray from the corresponding voxel (x,
y, z), and h.sub.phase is a weighting function relating to the time
difference of the projection P(.theta.+k.pi., {tilde over
(.beta.)}, q) or P(.theta.+k.pi., {tilde over (p)},q) from a
movement state, C.sub.R(k) representing a time position which
determines the periodic movement state of an examination area,
preferably an assigned heart cycle.
16. The method as claimed in claim 15, wherein the sum 16 P x , y ,
z ( ) = 1 H k q h z ( d x , y , z ( + k , { p ~ ~ } , q ) ) h phase
( t ( + k ) - c R ( k ) ) P ( + k , { p ~ ~ } , q ) which is
normalized to the sum H of the weights h 17 H = k q h z ( d x , y ,
z ( + k , { p ~ ~ } , q ) ) h phase ( t ( + k ) - c R ( k ) ) is
formed during the back-projection of the parallel data.
17. The method as claimed in claim 2, wherein the focal track is a
circular track.
18. The method as claimed in claim 2, wherein the focal track is a
spiral track, obtained by moving the focus on the circular track
about the system axis and simultaneously carrying out a relative
movement between the focus and the object to be examined in the
direction of the system axis.
19. The method as claimed in claim 5, wherein the focal track is a
circular track.
20. The method as claimed in claim 5, wherein the focal track is a
spiral track, obtained by moving the focus on the circular track
about the system axis and simultaneously carrying out a relative
movement between the focus and the object to be examined in the
direction of the system axis.
21. The CT device of claim 10, wherein the CT device is for
generating CT images of a heart region of a living being.
22. The CT device of claim 10, wherein the CT device is for
generating CT images of a heart region of a patient.
23. A method for generating CT images, comprising: scanning a
periodically movable object to be examined with a beam of rays
coming from at least one focus; detecting the rays using a
two-dimensionally designed detector array with a multiplicity of
distributed detector elements, wherein the detector elements of the
detector array are adapted to deliver output data representing an
attenuation of the rays when passed through the periodically
movable object; simultaneously collecting movement data upon
movement of the movable object, to enable assigning of detector
data and data resulting therefrom to movement states; filtering the
output data; three-dimensionally back-projecting the filtered
output data to generate at least one sectional image of a layer of
the movable object having a layer thickness, wherein each sectional
image represents absorption values, obtained from the output data,
of a voxel belonging to the layer of the movable object for the
radiation of the beam of rays, wherein a weighting function, for
weighting a spatial distance of a ray in question from a voxel in
question, is used for the back-projecting, and wherein a weighting
function, representing a time difference from an examination area
movement state to be represented is also used for the
back-projecting.
24. The method as claimed in claim 23, wherein the filtering is
carried out in the direction of a tangent to the focal track
belonging to the respective focal position.
Description
[0001] The present application hereby claims priority under 35
U.S.C. .sctn.119 on German patent application number DE 102 44
180.4 filed Sep. 23, 2002, the entire contents of which are hereby
incorporated herein by reference.
FIELD OF THE INVENTION
[0002] The invention generally relates to a method for imaging in
computer tomography. Preferably, it relates to one in which a
periodically moved object to be examined is scanned with the aid of
a beam of rays coming from a focus and a two-dimensionally designed
detector array, wherein the output data which are determined are
filtered in a suitable way and back-projected in order to obtain at
least one sectional image which represents the absorption values of
the section of the object to be examined in a particular movement
state. The invention furthermore generally relates to a CT device;
preferably one which is suitable for carrying out the method.
Background of the Invention
[0003] Methods for static objects have in principle been disclosed
under the term "3-D filtered back-projection". A disadvantage with
these methods is that redundant data which result from spiral scans
with a small incremental table feed, owing to multiple exposure of
the same voxel, are not used. The consequence of this is that the
radiation dose administered to the object to be examined is used
only incompletely for the imaging. There are also problems in
connection with carrying out image reconstruction methods in such a
way that a large number of provisional images are calculated by "2D
filtered back-projection" from output data which come from focal
track sections that are insufficient for the image reconstruction
per se; only in a second step are the provisional images
reformatted to form a definitive sectional image. These 2D methods
are less useful for detector arrays with a large width, i.e. a
large extent in the direction of the system axis, because an
extremely large number of provisional images then need to be
processed, which is problematic even when a large computing power
is available.
[0004] Cardio CT devices are in principle also generally known
which produce sectional recordings of periodically moved objects to
be examined, such as the human heart, by registering the movement
phases of the heart and by only using data which correlate with
particular movement phases or movement states to produce the
sectional images. A disadvantage with the known cardio CT devices,
however, is that they are only usable for detectors with a
comparatively small number of rows, or a small extent of the
detector in the system axis direction, since they do not take into
account the conical ray profile in the object to be examined.
[0005] Reference will be made to the Applicant's German patent
application DE 101 59 927.7, not yet published at the priority date
of the present application and the contents of which are
incorporated by reference herein in their entirety, which describes
a similar method and CT device but which relates only to the
imaging of static objects to be examined. The disclosure content of
this application is fully included here.
SUMMARY OF THE INVENTION
[0006] It is an object of an embodiment of the invention to provide
a method which, on the basis of 3D back-projection, makes it
possible to increase the image quality and which can utilize
redundant data of periodically moved objects to be examined with
the correct phase to form sectional images.
[0007] According to an embodiment of the invention, an object may
be achieved by a method and a CT device. Advantageous refinements
are described in the detailed description.
[0008] Accordingly, the inventor provides an embodiment of a method
for generating CT images of a periodically moving examination area,
preferably the heart region of a living being, preferably of a
patient, with at least the following method steps:
[0009] in order to scan a periodically moved object to be examined
with a beam of rays coming from at least one focus and with a
two-dimensionally designed detector array with a multiplicity of
distributed detector elements for detecting the rays of the beam of
rays, at least one focus is moved relative to the object to be
examined on at least one focal track circumscribing the object to
be examined with the detector array lying opposite, the detector
elements of the detector array delivering output data which
represent the attenuation of the rays when they pass through the
object to be examined,
[0010] movement data are simultaneously collected from the
periodically moving examination area, in order to be able to assign
the detector data and data resulting therefrom to the movement
states,
[0011] the output data are filtered,
[0012] the filtered output data are three-dimensionally
back-projected in order to generate at least one sectional image of
a layer of the object to be examined having a layer thickness,
[0013] each sectional image representing absorption values,
obtained from the output data, of the voxel belonging to the layer
of the object to be examined for the radiation of the beam of
rays,
[0014] on the one hand, a weighting function which weights the
spatial distance of a ray in question from the voxel in question
being used for the back-projection, and
[0015] on the other hand, a weighting function which represents the
time difference from the examination area movement state to be
represented in each case also being used for the
back-projection.
[0016] In a preferred embodiment of the method, the filtering is
carried out in the direction of the tangent to the focal track
belonging to the respective focal position. It has been shown that
a particularly high image quality can be achieved with this
filtering. The choice of this filter direction is based on the fact
that the described 2D method based on provisional images, which
allows a high image quality, could so to speak be converted into a
3D method if the focal track sections on which the calculation of
provisional images is based were to be shortened so much that they
can then be covered in a single projection, whose data are then
filtered in the direction of the tangent to the focal track, and
that it may be expected that such a 3D method would then allow as
good an image quality compared with the 2D method.
[0017] The method according to an embodiment of the invention can
then be implemented in a particularly straightforward way if a
conversion of the output data obtained in the form of rays
P(.alpha.,.beta.,q) in a fan-ray geometry into parallel data
available in the form of rays P(.theta.,.beta.,q) ("azimuthal
rebinning") or P(.theta.,p,q) ("complete rebinning", i.e. azimuthal
and radial "rebinning") in a parallel-ray geometry is carried out
before the filtering. In this case, referring to FIG. 3,
[0018] .alpha. is the focal angle,
[0019] .beta. is the fan angle,
[0020] q is the row index of the detector system corresponding to
the z coordinate,
[0021] .theta.=.alpha.+.beta. is the parallel fan angle,
[0022] p=R.sub.F sin(.beta.) is the parallel coordinate
corresponding to the ray distance from the rotation axis (system
axis), and
[0023] R.sub.F is the radius of the focal track.
[0024] According to a preferred embodiment of the invention,
back-projection of the parallel data is carried out in such a way
that the sum 1 P x , y , z ( ) = k q h z ( d x , y , z ( + k , { p
~ ~ } , q ) ) h phase ( t ( + k ) - c R ( k ) ) P ( + k , { p ~ ~ }
, q )
[0025] is formed in the course of the back-projection for each
voxel (x,y,z) for each .theta..epsilon.[0,.pi.[ for the rays
P(.theta.+k.pi., {tilde over (.beta.)},q) or P(.theta.+k.pi.,
{tilde over (p)}q) whose projection along the system axis passes
through (x,y), where
[0026] x,y,z are the coordinates of the respective voxel,
[0027] k is an integer corresponding to the number of half-circuits
of the focus which are involved in the reconstruction,
[0028] {tilde over (p)} are the parallel coordinates of those rays
whose projections along the system axis pass through the
coordinates (x,y) of the respective voxel (x,y,z),
[0029] {tilde over (.beta.)} are the fan angles of those rays whose
projections along the system axis pass through the coordinates
(x,y) of the respective voxel (x,y,z),
[0030] h.sub.z is a weighting function determining the layer
thickness of the layer of the object to be examined which is
represented in the sectional image being produced,
[0031] d is a function which is equal to the distance of the
respective ray from the corresponding voxel (x,y,z) or depends on
the distance of the respective ray from the corresponding voxel
(x,y,z), and
[0032] h.sub.phase is a weighting function relating to the time
difference t of the projection P(.theta.+k.pi.,{tilde over
(.beta.)},q) or P(.theta.+k.pi.,{tilde over (p)},q) from a movement
state,
[0033] c.sub.R (k) representing a time position which determines
the periodic movement state of an examination area, preferably an
assigned heart cycle.
[0034] The notation 2 { p ~ ~ }
[0035] here expresses the fact that the summation may be carried
out selectively for rays obtained by azimuthal "rebinning" or by
complete "rebinning", the filtering tangentially to the focal track
being filtering in the .beta. direction in the case of azimuthal
"rebinning" and filtering in the p direction in the case of
complete "rebinning".
[0036] The summation over both k and q then guarantees that all
rays passing through the same voxel are taken into account, and the
radiation dose delivered to the object to be examined is therefore
fully utilized.
[0037] In a particularly preferred embodiment, the sum 3 P x , y ,
z ( ) = 1 H k q h z ( d x , y , z ( + k , { p ~ ~ } , q ) ) h phase
( t ( + k ) - c R ( k ) ) P ( + k , { p ~ ~ } , q )
[0038] which is normalized to the sum H of the weights h 4 H = k q
h z ( d x , y , z ( + k , { p ~ ~ } , q ) ) h phase ( t ( + k ) - c
R ( k ) )
[0039] is formed during the back-projection of the parallel data.
This procedure allows a further improved image quality, since
possible over-emphasis of voxels which have received more rays than
other voxels is eliminated and corresponding artifacts are hence
avoided. The CT value of the respective voxel is obtained by
summation over .theta..
[0040] The method according to an embodiment of the invention may
for example be employed when, according to a variant of an
embodiment of the invention, the focal track is a circular track
(tomogram scan). According to a preferred variant of an embodiment
of the invention, however, the focal track is a spiral track which
is obtained by moving the focus on the circular track about the
system axis and simultaneously carrying out a relative movement
between the focus and the object to be examined in the direction of
the system axis. Larger volumes of the periodically moved object to
be examined may also readily be determined on the basis of such
spiral scanning.
[0041] In the case of a tomogram scan, k=1 or k=2 normally applies
for k. In the case of a spiral scan, k will be selected so that all
rays in the projection angle .theta. belonging to the parallel
coordinate p or to the fan angle .beta. whose distance d.sub.x,y,z
does not exceed a suitable threshold are recorded for the image
reconstruction.
[0042] The present invention will become more fully understood from
the detailed description of preferred embodiments given hereinbelow
and the accompanying drawings which are given by way of
illustration only, and thus are not limitative of the present
invention, and wherein:
[0043] FIG. 1 illustrates a CT device having a plurality of rows of
detector elements in a partly perspective and partly
block-diagrammatic representation;
[0044] FIG. 2 illustrates a longitudinal section through the device
according to FIG. 1;
[0045] FIG. 3 illustrates a view illustrating the "rebinning";
[0046] FIG. 4 illustrates a typical ECG of a human heart;
[0047] FIG. 5 illustrates a movement state of the heart,
represented by relative volume change and correlated with the ECG
in FIG. 4;
[0048] FIG. 6 illustrates the time profile of the weighting
function h.sub.phase correlated with FIGS. 4 and 5.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0049] FIGS. 1 and 2 represent a 3rd generation CT device suitable
for carrying out the method according to an embodiment of the
invention. Its measuring arrangement, denoted overall by 1, has an
x-ray source, denoted overall by 2, with a source-side beam
diaphragm 3 (FIG. 2) placed in front of it and a detector system 5
designed as a flat array of a plurality of rows and columns of
detector elements--one of these is denoted by 4 in FIG. 1 --with a
detector-side beam diaphragm 6 (FIG. 2) placed in front of it. Only
eight rows of detector elements 4 are represented in FIG. 1 for the
sake of clarity. As indicated by dots in FIG. 2, however, the
detector system 5 has further rows of detector elements 4.
[0050] The x-ray source 2 with the beam diaphragm 3, on the one
hand, and the detector system 5 with the beam diaphragm 6, on the
other hand, as can be seen in FIG. 2, are fitted opposite one
another on a rotary frame 7 in such a way that a pyramidal x-ray
beam whose edge rays are denoted by 8, which comes from the x-ray
source 2 during operation of the CT device and is stopped down by
the adjustable beam diaphragm 3, strikes the detector system 5. In
this case, the beam diaphragm 6 is adjusted to correspond with the
x-ray beam cross section set by means of the beam diaphragm 3, in
such a way as to expose only the area of the detector system 5
which can receive the x-ray beam directly. In the operating mode
illustrated in FIGS. 1 and 2, these are eight rows of detector
elements 4, which will be referred to below as the active rows. The
further rows indicated by dots are covered by the beam diaphragm 6
and are therefore inactive.
[0051] Each row of detector elements 4 has a number K of detector
elements, .beta..sub.k=.beta..sub.l to .beta..sub.K being the
channel index and a fan angle .beta..sub.k being assigned to each
detector element. The fan angle of the central detector element is
equal to zero; the fan angles of the two outermost detector
elements are .beta..sub.l=+.beta..sub.max and
.beta..sub.K=-.sub.max.
[0052] The active rows L.sub.q of detector elements 4 are denoted
by L.sub.l to L.sub.Q in FIG. 2, q=1 to Q being the row index,
which corresponds so to speak to the z coordinate in the case of
the exemplary embodiment being described.
[0053] The x-ray beam has the cone angle .phi. indicated in FIGS. 1
and 2, which is the aperture angle of the x-ray beam in a plane
containing the system axis Z and the focus F. The aperture angle of
the x-ray beam in a plane perpendicular to the system axis Z and
containing the focus F (fan aperture angle) is 2.beta..sub.max and
is indicated in FIG. 1.
[0054] The rotary frame 7 can be set in rotation about a system
axis denoted by Z, by means of a drive instrument 22. The system
axis Z extends parallel to the z axis of a right-angled spatial
coordinate system represented in FIG. 1.
[0055] The columns of the detector system 5 also extend in the
direction of the z axis while the rows, whose width b is measured
in the direction of the z axis and is for example 1 mm, extend
transversely to the system axis Z or the z axis.
[0056] In order to be able to bring an object to be examined, for
example a patient, into the beam path of the x-ray beam, a support
device 9 is provided which can be displaced parallel to the system
axis Z, i.e. in the direction of the z axis, so as to obtain
synchronization between the rotation movement of the rotary frame 7
and the translation movement of the support device such that the
ratio of translation speed to rotation speed is constant, this
ratio being adjustable by selecting the intended value of the
incremental feed .nu. of the support device per revolution of the
rotary frame.
[0057] It is hence possible for a volume of an object to be
examined, lying on the support device 9, to be examined in the
course of a volume scan; the volume scan may be carried out in the
form of a spiral scan such that a multiplicity of projections from
different projection directions are recorded by means of the
measuring unit per revolution of the measuring unit 1, by
simultaneous rotation of the measuring unit 1 and translation of
the support device 9. During the spiral scan, the focus F of the
x-ray source moves relative to the support device 9 on a spiral
track denoted by S in FIG. 1. The spiral scan must extend over at
least .pi.+2.beta..sub.max in the .alpha. direction, in order to
permit the complete reconstruction of one CT image per row of
detector elements, although it may be arbitrarily longer within the
technical limitations of the CT device.
[0058] Owing to the fact that there are a plurality of rows of
detector elements 4, a volume of the object to be examined may
nevertheless also be examined in the course of a so-called tomogram
scan, in which no relative movement in the direction of the z axis
takes place between the measuring unit 1 and the support device 9
(.nu.=0). In the case of the tomogram scan, the size of the
examined volume is hence determined by the number of active rows of
detector elements 4. During a tomogram scan, the focus F moves on a
circular focal track, which lies in a plane referred to below as
the mid-plane.
[0059] The tomogram scan may be carried out in the form of a
part-circuit or in the form of the full circuit, the part-circuit
including a part-circuit interval of at least .pi.+2.beta..sub.max
(one half-circuit plus the fan aperture angle) which permits the
complete reconstruction of a CT image, while a full scan comprises
2.pi..
[0060] The measurement data read out in parallel from the detector
elements of each active row of the detector system 5 during the
spiral or tomogram scan, which correspond to the individual
projections P(.alpha., .beta.,q) in a fan-ray geometry, are
subjected to digital/analog conversion in a data processing unit
10, serialized and transmitted to an image computer 11.
[0061] After preprocessing of the measurement data in a
preprocessing unit 12 of the image computer 11, the resulting
datastream is sent to a sectional image reconstruction unit 13
which, from the measurement data, reconstructs sectional images of
intended layers of the object to be examined on the basis of
"filtered back-projection" by using a method according to an
embodiment of the invention which is yet to be described in
detail.
[0062] The CT images are made up of matricially assembled pixels
(pixel=picture element), the pixels being assigned to the
respective image plane, each pixel being assigned a CT number in
Hounsfield units (HU) and the individual pixels being represented
according to a CT-number/gray-value scale in a gray value
corresponding to its respective CT number. Each pixel in this case
illustrates a voxel (voxel=volume element) of the layer of the
object to be examined illustrated in the CT image. Since
measurement data corresponding to a plurality of layers of the
object to be examined are available owing to the multiple-row
nature of the detector system 5 and, where applicable, the spiral
scan, 3D data are available which are subjected to 3D
back-projection in the scope of an embodiment of the invention. In
the end, 3D image data are available in the form of a
three-dimensional matrix, for example with the axes x,y,z, each
element of the matrix corresponding to one voxel (x,y,z) and
containing the gray value corresponding to the associated CT
number. Those elements of the three-dimensional matrix which have
the same x,y or z value then respectively constitute a planar
sectional image of the layer of the object to be examined
corresponding to the relevant x,y or z value.
[0063] The images reconstructed by the sectional image
reconstruction unit 13 are represented on a display unit 16
connected to the image computer 11, for example a monitor.
[0064] The x-ray source 2, for example an x-ray tube, is supplied
by a generator 13 with the required voltages and currents, for
example the tube voltage U . In order to be able to set them to the
respectively required values, the generator unit 17 is assigned a
control unit 18 with a keyboard 19, which allows the necessary
settings to be made.
[0065] The other operation and control of the CT device is also
carried out by way of the control unit 18 and the keyboard 19,
which is illustrated by the fact that the control unit 18 is
connected to the image computer 11.
[0066] Inter alia, it is possible to set the number Q of active
rows of detector elements 4 and therefore the position of the beam
diaphragms 3 and 6, for which the control unit 18 is connected to
adjustment units 20 and 21 assigned to the beam diaphragms 3 and 6.
It is also possible to set the rotation time .tau. which the rotary
frame 7 requires for a full revolution, which is illustrated by the
fact that the drive unit 22 assigned to the rotary frame 7 is
connected to the control unit 18.
[0067] Although it is also possible in principle to carry out the
method according to an embodiment of the invention in a fan-ray
geometry, the described CT device is preferably operated in a mode
in which the method according to an embodiment of the invention is
carried out in a parallel-ray geometry.
[0068] Accordingly, the data obtained in a fan-ray geometry during
the scan of the patient's body area relevant to the respective
examination by spiral or tomogram scanning are first converted into
a parallel-ray geometry, in a manner which is known per se, by a
method generally referred to as "rebinning". This conversion is
based on re-sorting of the data obtained in a fan-ray geometry, in
such a way that rays from different projections recorded in a
fan-ray geometry are taken and combined to form a projection in a
parallel-ray geometry. In a parallel-ray geometry, data from an
interval with the length .pi. are sufficient in order to be able to
reconstruct a complete image. In order to be able to obtain these
data, data in a fan-ray geometry from an interval with the length
.pi.+2 .beta..sub.max must nonetheless be available.
[0069] A projection in a parallel-ray geometry is illustrated in
FIG. 3. Accordingly, all n parallel rays RP.sub.1 to RP.sub.N of
this projection assume the parallel fan angle .theta. with respect
to the x axis of the coordinate system represented in FIG. 3 and
corresponding to that according to FIG. 1.
[0070] The conversion from a fan-ray geometry to a parallel-ray
geometry will be explained below with the aid of the parallel ray
RP, represented by a bold line in FIG. 3.
[0071] The parallel ray RP.sub.1 originates from a projection
obtained in a fan-ray geometry for the focal position F.sub.1 lying
on the focal track S. The central ray RF.sub.z1i belonging to this
projection in a fan-ray geometry and passing through the rotation
axis 14, and therefore the z axis of the coordinate system, is also
indicated in FIG. 3. The focal position F.sub.1 corresponds to the
focal angle .alpha..sub.1; this is the angle formed between the x
axis and the central ray RF.sub.z1. With respect to the central ray
RF.sub.z1, the ray RP.sub.1 has the fan angle .beta.. It is
therefore easy to see that the following applies for the parallel
fan angle .theta.: .theta.=.alpha.+.beta..
[0072] The ray distance p from the rotation axis 14 or the z axis,
measured at right angles to the respective parallel ray, is given
by p=R.sub.F sin(.beta.).
[0073] As is made clear by the central ray RP, represented by an
emboldened line in FIG. 3 and passing through the rotation axis 14
or the X axis, this ray is the central ray of a projection in a
fan-ray geometry recorded in a fan geometry for the focal position
F.sub.z at the focal angle .alpha..sub.z. Since .beta.=0 applies
for the central ray of a projection recorded in a fan-ray geometry,
it is clear that the following applies for the case of central
rays:
[0074] Depending on whether azimuthal or complete "rebinning" is
carried out, the parallel projections are in the form
P(.alpha.,.beta.,q) or in the form P(.theta., p,q), where
[0075] .alpha. is the focal angle,
[0076] .beta. is the fan angle,
[0077] q is the row index of the detector system corresponding to
the z coordinate,
[0078] .theta.=.alpha.+.beta. is the parallel fan angle,
[0079] p=R.sub.F sin(.beta.) is the parallel coordinate
corresponding to the ray distance from the rotation axis (system
axis), and
[0080] R.sub.F is the radius of the focal track.
[0081] In a first operating mode corresponding to a first
embodiment of the method according to the invention, which can be
selected by means of the keyboard 19, the described CT device works
on the basis of projections obtained by azimuthal "rebinning". The
data corresponding to these projections are filtered in the .beta.
direction in the case of azimuthal "rebinning", i.e. respectively
in the direction of the tangent T belonging to the focal position
of the central ray of the respective parallel projection (see FIG.
3), specifically by using one of the filter kernels which are
conventional in computer tomography, for example the Shepp-Logan or
Ramachandran-Lakshminarayanan kernel.
[0082] The parallel data filtered in this way are then
back-projected in such a way that the sum 5 P x , y , z ( ) = k q h
z ( d x , y , z ( + k , ~ , q ) ) h phase ( t ( + k ) - c R ( k ) )
P ( + k , ~ , q )
[0083] is formed in the course of the back-projection for each
voxel (x,y,z) for each .theta..epsilon.[0, .pi.[ for the rays
P(.theta.+k.pi.,{tilde over (.beta.)}, q) whose projection along
the system axis passes through (x,y), where
[0084] x,y,z are the coordinates of the respective voxel,
[0085] k is an integer corresponding to the number of half-circuits
of the focus which are involved in the reconstruction,
[0086] {tilde over (.beta.)} are the fan angles of those rays whose
projections along the system axis pass through the coordinates
(x,y) of the respective voxel (x,y,z),
[0087] h.sub.z is a weighting function determining the layer
thickness of the layer of the object to be examined which is
represented in the sectional image being produced,
[0088] d is a function which is equal to the distance of the
respective ray from the corresponding voxel (x,y,z) or depends on
the distance of the respective ray from the corresponding voxel
(x,y,z), and
[0089] h.sub.phase is a weighting function relating to the time
difference t of the projection P(.theta.+k.pi.,{tilde over
(.beta.)},q) P(.theta.+k.pi.,{tilde over (p)}, q) from a movement
state,
[0090] c.sub.R(k) representing a time position which determines the
periodic movement state of an examination area, preferably an
assigned heart cycle.
[0091] Owing to the chosen filter direction and owing to the
summation over both k and q, on the one hand "cone beam artifacts"
are avoided and, on the other hand, all rays passing through a
voxel (x,y,z) are taken into account with a view to high dose
utilization. Furthermore, only rays which are correlated with that
movement state of the object to be examined which is intended to be
observed are taken into account by the weighting function
h.sub.phase, and measurements not correlated with it are
suppressed.
[0092] The absorption value .mu..sub.x,y,z assigned to a voxel
x,y,z is obtained by summation over .theta. over at least one
half-circuit, i.e. by forming 6 x , y , z = P x , y , z ( ) .
[0093] The CT number corresponding to the respective absorption
value is determined from the absorption value in a conventional
way.
[0094] In this case, various weighting functions h and various
functions d can be set by means of the keyboard 19.
[0095] For example, a triangle or trapezium function may be
suitable as the weighting function h.
[0096] The distance of the respective parallel ray from the voxel
x,y,z or, instead of this, for example the z (axial) component of
this distance may be set as the function d.
[0097] In a variant of the first operating mode described above,
the sum 7 P x , y , z ( ) = 1 H k q h z ( d x , y , z ( + k , ~ , q
) ) h phase ( t ( + k ) - c R ( k ) ) P ( + k , ~ , q )
[0098] which is normalized to the sum H of the weights h.sub.z and
h.sub.phase 8 H = k q h z ( d x , y , z ( + k , ~ , q ) ) h phase (
t ( + k ) - c R ( k ) )
[0099] is formed for the back-projection of the parallel data. This
permits a further improved image quality since possible
over-emphasis of voxels which are "illuminated", i.e. struck by
rays, in a plurality of half-circuits is eliminated and
corresponding artifacts are hence avoided. This redundancy occurs
in spiral scans whenever the relative displacement taking place per
full circuit of the measuring arrangement is so small (small pitch)
that voxels are irradiated several times.
[0100] An operating mode corresponding to another embodiment of the
method according to the invention, which can be selected via the
keyboard 19, differs from the first operating mode in that the
described CT device works not on the basis of projections obtained
by azimuthal "rebinning", but on the basis of projections obtained
by complete "rebinning". The data corresponding to these
projections are filtered in the p direction in the case of complete
"rebinning", i.e. likewise respectively in the direction of the
tangent T belonging to the focal position of the central ray of the
respective parallel projection (see FIG. 3).
[0101] Accordingly, the sum 9 P x , y , z ( ) = k q h z ( d x , y ,
z ( + k , p ~ , q ) ) h phase ( t ( + k ) - c R ( k ) ) P ( + k , p
~ , q )
[0102] is formed in the course of the back-projection for the
parallel data filtered in this way, {tilde over (p)} being the
parallel coordinates of those rays whose projections along the
system axis pass through the coordinates (x,y) of the respective
voxel (x,y,z).
[0103] Also in the case of the second operating mode, in a variant,
a sum which is normalized to the sum H of the weights h.sub.z and
h.sub.phase 10 H = k q h z ( d x , y , z ( + k , p ~ , q ) ) h
phase ( t ( + k ) - c R ( k ) )
[0104] is formed for the back-projection of the parallel data,
namely the sum 11 P x , y , z ( ) = k q h z ( d x , y , z ( + k , p
~ , q ) ) h phase ( t ( + k ) - c R ( k ) ) P ( + k , p ~ , q )
.
[0105] In the case of the first and second operating modes
described above, a functionality of the method according to an
embodiment of the invention is provided in which all rays whose
projection along the rotation axis 14 or the z axis passes through
x,y are taken into account in relation to a voxel (x,y,z). How and
to what extent these rays are taken into account will be determined
by the weighting function h.sub.z and the function d.
[0106] FIGS. 4 to 6 show the time correlation between a patient's
schematically represented ECG recording in FIG. 4, with the
well-known typical profile over the atrial part with P wave and PQ
interval, followed by the ventricular part with the QRS complex, ST
interval, T wave and U wave. In this representation, only the P
wave, QRS complex and T wave are highlighted as characteristic.
[0107] This typical and periodic ECG profile is correlated with
particular movement states of the heart, which are represented in
FIG. 5 with synchronous timing by plotting the relative volume
change V/V.sub.0(t) of the heart. If this profile is roughly
divided into two phases, then a movement phase I and a resting
phase II are found. According to an embodiment of the invention,
the CT recordings should respectively correspond to a particular
phase so that, according to the respective phases, the weighting
function h.sub.phase whose profile is represented in FIG. 6
guarantees that only information pertaining to this particular
movement situation is in fact used for the imaging. Since it may be
assumed that only minor movement takes place over a relatively long
time span, namely the resting phase II of the heart, the entire
resting phase II may, for example, be weighted heavily while the
movement phase I is weighted lightly, as can be seen from FIG. 6.
The bold line in FIG. 6 shows such a profile of the weighting
function h.sub.phase with only two different values 1 and 0.
[0108] Alternatively, however, it is also possible to take into
account the actual relative volume change and therefore the
movement state in the heart, if, as represented in the profile of
the weighting function h.sub.phase represented by dashes, different
and more finely differentiated weightings are introduced depending
on the extent of the resting of the heart. Furthermore, a weighting
function which decreases with the time difference from a
predetermined movement state may be chosen. This is represented,
for example, by the dotted profile of the weighting function.
[0109] The CT device, however, may also have other operating modes
selectable via the keyboard 19, which correspond to those described
above with the difference that the theoretical ray passing through
the respective voxel (x,y,z) is determined for a given focal
position, and then only those rays which can actually make a
contribution to the sum, both with reference to their position and
with reference to the movement state of the relevant object to be
examined which is of interest, are included in the summation in the
course of the back-projection by taking into account the weighting
functions h.sub.z and h.sub.phae and the function d.
[0110] In the case of the described exemplary embodiments, the
relative movement between the measuring unit 1 and the support
device 9 is respectively achieved by displacing the support device
9. It is, however, also possible in the scope of the invention to
leave the support device 9 positionally fixed and to displace the
measuring unit 1 instead of it. It is furthermore possible in the
scope of the invention to generate the required relative movement
by displacing both the measuring unit 1 and the support device
9.
[0111] The conical x-ray beam has a rectangular cross section in
the case of the described exemplary embodiment. Other
cross-sectional geometries, however, are also possible in the scope
of the invention.
[0112] Third-generation CT devices are employed in connection with
the exemplary embodiments described above, i.e. the x-ray source
and the detector system are moved together about the system axis
during the imaging. The invention, however, may also be employed in
connection with fourth-generation CT devices, in which only the
x-ray source is moved about the system axis and it interacts with a
stationary detector ring, so long as the detector system is a
multiple-row array of detector elements.
[0113] The method according to an embodiment of the invention may
also be employed in fifth-generation CT devices, i.e. CT devices in
which the x-radiation does not come from just one focus, but from a
plurality of foci of one or more x-ray sources which are moved
about the system axis, so long as the detector system has a
multiple-row array of detector elements.
[0114] The CT devices used in connection with the exemplary
embodiments described above have a detector system with detector
elements arranged in the manner of an orthogonal matrix. The
invention, however, may also be employed in connection with CT
devices whose detector system has a flat-arranged array with
detector elements in a different way.
[0115] The exemplary embodiments described above relate to the
medical use of the method according to the invention. The
invention, however, may also be used outside medicine, for example
for the material examination of periodically moving machines or
machine elements.
[0116] Exemplary embodiments being thus described, it will be
obvious that the same may be varied in many ways. Such variations
are not to be regarded as a departure from the spirit and scope of
the present invention, and all such modifications as would be
obvious to one skilled in the art are intended to be included
within the scope of the following claims.
* * * * *